.png){kind=icon}
.png){kind=icon}
Return
Structured Covariance Tensor Reconstruction via T-Product for 2-D DOA Estimation of Coherent Signals
.png){kind=icon}
PRE
AI
Q
L
Z
DOI：10.1109/taes.2026.3712332.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This work proposes a high-resolution direction-of-arrival (DOA) estimation method for coherent signal environments, utilizing a uniform rectangular array. First, by exploiting the rank-one structure of fully coherent signals along the third mode, a structure-matched subspace projection is introduced to preprocess the observation tensor, such that dominant signal components are preserved while noise components orthogonal to the signal subspace are effectively suppressed. Second, under the tensor t-product framework, a third-order covariance tensor is constructed from the projected signal model to capture the multidimensional dependencies and convolutional correlations embedded in the array observations more effectively. Third, we propose a structured tensor reconstruction scheme, in which selected slices of the coherent covariance tensor are reorganized to generate a decorrelated covariance representation. The reconstructed tensor is further demonstrated to possess a decorrelated canonical polyadic (CP) structure, thereby facilitating accurate parameter estimation through canonical polyadic decomposition. In this way, the proposed method jointly exploits subspace denoising, tensor covariance modeling, and structured reconstruction within a unified framework. Comparative simulations confirm that the newly proposed method exhibits superior 2D-DOA estimation capabilities for coherent sources, relative to conventional spatial smoothing and matrix reconstruction approaches. Most significantly, this performance edge remains robust even under challenging low-SNR operating conditions.
Keywords:
Coherent signals
direction-of-arrival (DOA) estimation
subspace projection
t-product
tensor reconstruction
Journal
IF:
5.7
Papers:
651
Citations:
2.4W
Organization
Cited Papers